lable at ScienceDirect

Polymer 51 (2010) 1088–1099

Contents lists avai

Polymer

journal homepage: www.elsevier .com/locate/polymer

Effects of molecular weight on poly(u-pentadecalactone) mechanicaland thermal properties

Jiali Cai a, Chen Liu a, Minmin Cai a, Jie Zhu b, Feng Zuo b, Benjamin S. Hsiao b,**, Richard A. Gross a,*

a NSF I/UCRC for Biocatalysis and Bioprocessing of Macromolecules, The Polytechnic Institute of New York University, Six Metrotech Center, Brooklyn, NY 11201, USAb Department of Chemistry, Stony Brook University, Stony Brook, NY 11794, USA

a r t i c l e i n f o

Article history:Received 17 November 2009Received in revised form4 January 2010Accepted 7 January 2010Available online 14 January 2010

Keywords:Poly(u-pentadecalactone)Molecular weightTensile

* Corresponding author. Tel.: þ1 718 260 3408; fax** Corresponding author.

E-mail address: jlcai2004@yahoo.com (R.A. Gross)

0032-3861/$ – see front matter � 2010 Elsevier Ltd.doi:10.1016/j.polymer.2010.01.007

a b s t r a c t

A series of poly(u-pentadecalactone) (PPDL) samples, synthesized by lipase catalysis, were prepared bysystematic variation of reaction time and water content. These samples possessed weight-averagemolecular weights (Mw), determined by multi-angle laser light scattering (MALLS), from 2.5� 104 to48.1� 104. Cold-drawing tensile tests at room temperature of PPDL samples with Mw between 4.5� 104

and 8.1� 104 showed a brittle-to-ductile transition. For PPDL with Mw of 8.1� 104, inter-fibrillar slippagedominates during deformation until fracture. Increasing Mw above 18.9� 104 resulted in enhancedentanglement network strength and strain-hardening. The high Mw samples also exhibited tough prop-erties with elongation at break about 650% and tensile strength about 60.8 MPa, comparable to linear highdensity polyethylene (HDPE). Relationships among molecular weight, Young’s modulus, stress, strain atyield, melting and crystallization enthalpy (by differential scanning calorimetry, DSC) and crystallinity(from wide-angle X-ray diffraction, WAXD) were correlated for PPDL samples. Similarities and differencesof linear HDPE and PPDL molecular weight dependence on their mechanical and thermal properties werealso compared.

� 2010 Elsevier Ltd. All rights reserved.

1. Introduction

Polyethylene is the most widely used commodity polymer. It isfound in many consumer products, such as milk jugs, detergentbottles, margarine tubs, garbage containers, water pipes, just toname a few. Poly(u-pentadecalactone) (PPDL) is a new type ofthermoplastic that can be synthesized by lipase catalysis [1–3]. Thechemical structure of PPDL, with 14 methylene groups and an in-chain ester linkage in each repeating unit, is very similar to that oflinear high density polyethylene (HDPE) (Scheme 1). Polyethylene(PE) cannot be easily decomposed into small molecules after usage.To achieve extensive degradation of the PE carbon backbone,treatment of PE with strong oxidized agents such as nitric acid [4],ozone [5] and permanganic acid [6] or pyrolysis at high reactiontemperatures (above 370 �C) [7,8] is required. Therefore, ‘‘whitepollution’’ [9,10] from un-recycled PE plastics is a mountingproblem that mankind must confront. An important advantage ofPPDL over PE is that the former has ester groups in the backbonethat are susceptive to chain breakage. Consequently, gentle

: þ1 718 260 3075.

.

All rights reserved.

enzymatic hydrolysis can, in principle, be used to decompose PPDLback into monomer building blocks. Currently, on-going studies arein progress to find a suitable enzyme for PPDL biological recyclingboth in our laboratory and elsewhere [11].

Chemical catalysts such as potassium alkoxides [12], diethylzinc[13,14] and yttrium isopropoxide [15] can be used for conversion ofmacrolactones to polyesters. For u-pentadecalactone (PDL), the useof immobilized lipase catalysis has been proven to be superior tochemical catalyzed routes for making polymers, resulting in morerapid polymerization kinetics as well as yielding polyesters of rela-tively higher molecular weight. Immobilized lipase-catalyzed poly-merization of macrolactones was first published by Uyama et al. [16].Our laboratory demonstrated that, using Novozym 435 that consistsof Candida antarctica lipase B (CALB) physically immobilized ona macroporous support, the polymerization process of PDL givesPPDL with number-average molecular weight (Mn) up to 8.6�104 inyields exceeding 90% (route is illustrated in Scheme 1) [17]. Thus,enzyme-catalyzed preparation of PPDL with high molecular weighthas provided suitable materials for evaluation of their physico-mechanical properties. For example, mechanical properties of PPDLwith Mn 6.5�104 (polydispersity, Mw/Mn, 2.0) [18] and thiol-func-tionalized PPDL telechelics [19] have been studied.

Previous crystallographic work [20] indicated that the a andb parameters of PPDL unit cells are slightly larger than those of PE’s

CH2

O

O

11

PDL

Enzymatic Catalysis Synthesis Route

Toluene

70°C

High density polyethylene

---CH2-CH2-CH2-CH2-CH2-CH2-CH2-CH2-CH2-CH2-CH2-CH2-CH2-CH2-CH2-CH2-(-CH2-CH2-)n---

Poly(ω-pentadecalactone)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

CH2-CH2-CH2-CH2 -CH2-CH2-CH2-CH2-CH2-CH2-CH2-CH2-CH2-CH2-C-O-CH2-CH2---O

( )n

CH2 CH2

a,Chromium/silica catalysts

b, Ziegler-natta catalysts

c,Metallocene catalysts

---

Scheme 1. Comparison of molecular structures of high density polyethylene (HDPE) and poly(u-pentadecalactone) obtained via enzyme-catalyzed ring-opening polymerizations.

J. Cai et al. / Polymer 51 (2010) 1088–1099 1089

and the until cell parameter along the fiber axis is much larger thanthat of PE’s. DMA tests showed that PPDL with Mn 6.5�104 hada glass transition at �27 �C. Also, the observed high storagemodulus was attributed to high crystallinity as determined by DSCand WAXS [18]. PPDL-based copolymers with other monomersincluding trimethylene carbonate [2], p-dioxanone [3] and u-cap-rolactone [21] were synthesized and studied by thermal and X-rayanalysis. Copolymers were found to be highly crystalline randomcopolymers over the entire composition range. This behavior wasattributed to co-crystallization of comonomer units in a commonlattice or isomorphic substitution of comonomer units.

The current study aimed to investigate the effect of PPDLmolecular weight on its mechanical, thermal and rheologicalproperties. The chosen synthetic methods enabled the preparationof PPDL with Mw values up to 48.1�104 and PDI values close to 2.0.Films were prepared by press-molding at 130 �C and tensile testingwas performed on these samples. Based on the shape of stress–strain curves, a brittle-to-ductile transition along with maximumelongation at break was observed. Since the chemical structure ofPPDL is similar to polyethylene (PE), its thermal and mechanicalproperties were compared with those of a commercially obtainedPE sample [22]. In addition, PPDL films with different molecularweights were analyzed by differential scanning calorimetry (DSC),wide-angle X-ray diffraction (WAXD), small-angle X-ray scattering(SAXS) and dynamic mechanical analysis (DMA) to reveal themolecular weight dependence of mechanical, thermal, crystallineand rheological properties. It was found that Young’s modulus andstress at break exhibit distinct crystallinity dependence; in addition,

differences, of elongation at break and true stress at break, betweenPPDL and linear PE, are discussed.

2. Experimental

2.1. Materials

Samples of u-pentadecalactone (PDL, 98%) and anhydrous p-xylene(>99%) were purchased from Aldrich Chemical Co. and were used asreceived. Chloroform was purchased from PHARMCO-AAPER Inc.(>99.9%). Anhydrous toluene (98%), purchased from Aldrich ChemicalCo., was dried over sodium and then was distilled under nitrogen.Novozym 435 (specific activity 10,000 PLU/g) was a gift from Novo-zymes (Bagsvaerd, Denmark) and consists of Candida antarctica LipaseB (CALB) physically adsorbed within the macroporous resin LewatitVPOC 1600 (poly[methyl methacrylateco-butyl methacrylate],supplied by Bayer).

2.2. Sample preparation

2.2.1. Enzyme-catalyzed synthesis of PPDL samples 1–3 (Table 1)Pentadecalactone (PDL, 40 g) was polymerized to prepare PPDL

samples 1–3 at 70 �C, with magnetic stirring, in toluene (mono-mer:toluene¼ 1:2 wt/v), using Novozym 435 as the catalyst forpredetermined reaction times (see Table 1). Reactions were carriedout in 500 mL round-bottom flasks fitted with a magnetic stirrerand an inlet/outlet for nitrogen gas. No precautions were taken todry the monomer, catalyst, solvent or glassware. The ratio of

Table 1PPDL Mw and PDI values for PPDL samples and corresponding reaction times forsample synthesis.

Sample number Mw� 10�4 a PDIb Reaction time (h)

1 2.5 4.9 62 4.5 3.3 133 8.1 3.1 244 18.9 1.6 85 28.0 2.0c 166 48.1 1.8 26

a From light scattering.b PDI is polydispersity index that was determined by GPC relative to polystyrene.c Calculation performed without inclusion of low molecular weight peaks

observed in the GPC chromatogram (see Fig. 7).

J. Cai et al. / Polymer 51 (2010) 1088–10991090

Novozym 435 to PDL was 1% (w/w). All liquid transfers were per-formed by syringe through rubber septum caps under nitrogen. Thefinal product mixture was dissolved in cold chloroform, filtered toremove catalyst beads, precipitated in methanol to removeunreacted PDL and then solvent was removed at 50 �C undervacuum for 16 h.

2.2.2. Enzyme-catalyzed synthesis of PPDL samples 4–6 (Table 1)The method used was a variation of that described by de Geus

[23] to obtain PPDL of higher molecular weight. PDL (50 g) waspolymerized to prepare PPDL samples 4–6 in toluene (monomer:-toluene¼ 1:2 w/v), using Novozym 435 as the catalyst for pre-determined reaction times (see Table 1). Reactions were conductedin 500 mL three-neck round-bottom flasks fitted with a glassoverhead stirrer and connected to nitrogen and vacuum lines viaa fire stone valve. The ratio of Novozym 435 to PDL was 1% (w/w).Vacuum pressure in the system was controlled (�0.2 mmHg) by aJ-KEM vacuum regulator. Novozym 435 was dried for 16 h at 50 �Cunder 2 mmHg. The reaction vessel containing 3 Å sieves was pre-dried at 150 �C (ambient pressure) for 16 h and, thereafter, wasflame dried and the atmosphere was switched to dry nitrogen priorto monomer and solvent addition. Glass syringes and needles werepre-dried at 150 �C for 16 h and were cooled under a nitrogen purgetill they was sufficiently cool to be handled for solvent andmonomer transfers. Dry toluene (80 g) PDL (40 g) was added to theflask by syringe under nitrogen. Then, one port on the vessel wasopened while purging with nitrogen and dried Novozym 435 wasadded. Reaction contents were heated by immersing the flask in anexternal oil bath at 85 �C. After the allotted reaction time, 150 mL ofpre-heated p-xylene (100 �C) was added to the highly viscousslurry of enzyme, PPDL and toluene. The slurry was stirred for 1 h todissolve PPDL and then filtered using a pre-heated Buchner-funnel.Small aliquots of pre-heated (100 �C) p-xylene solvent were used toquantitatively transfer reaction products. The filtrate was addedslowly to cold methanol (200 mL) to precipitate PPDL that wascollected using a Buchner-funnel. Finally, the polymer was driedunder vacuum (2 mmHg) for 16 h at 50 �C to remove solvents.

2.3. Nuclear magnetic resonance characterization

Both 1H and 13C NMR spectra were recorded at room tempera-ture on a DPX300 spectrometer (Bruker Instruments, Inc.) at300 MHz in chloroform-d. Chemical shifts (in parts per million) for1H and 13C NMR spectra were referenced relative to tetramethylsi-lane as an internal reference at 0.00. All synthesized PPDL materialsdescribed herein had identical NMR spectra with signals andassignments as follows: –[C(]O)–CH2

b–CH2c–(CH2

d–CH2d–)5–CH2

c–CH2

a–O]– 1H NMR (CDCl3, d): 4.01 (t, J 6.5 Hz, CH2aO); 3.58 (t, J 6.5 Hz,

CH2OH); 2.24 (t, J 7.5 Hz, CH2bCO); 1.59, 1.22 (brs, CH2

c,d) ppm. 13CNMR (CDCl3, d): 173.9 (COCH2), 64.4(CH2

aO), 34.4(OCOCH2b),

29.6–29.1, 28.6, 25.9, 25.0 (all other carbons) ppm.

2.4. Molecular weight determination

To determine the absolute molecular weight of PPDL samples,a Wyatt HELLOS multi-angle light scattering detector and a WyattOptilab rEx differential refractive index detector were used. Thesetwo detectors were connected by a steel tube with inner diameter0.5 mm and a CRZEL syringe pump. The solution concentrationranged from 5�10�4 to 5�10�3 g/mL. Astra V software was usedto acquire and process the data according to Zimm plots.

The relative molecular weight of PPDL samples were deter-mined by gel permeation chromatography (GPC) using a WatersHPLC system equipped with a model 510 pump, model 717 auto-sampler, and model 410 refractive index detector with 500, 103, 104,and 105 Å Ultrastyragel columns in series. Waters Empower GPCsoftware (Version 3, Viscotek Corp.) was used for data analysis.Chloroform was used as eluent at a flow rate of 1.0 mL/min. Sampleconcentrations and injection volumes were 0.2% w/v and 100 mL,respectively. The number-average molecular weight (Mn) andweight-average molecular weight (Mw) were determined based ona calibration curve generated by narrow molecular weight poly-styrene standards (Aldrich Chemical Co). GPC and light scatteringdetermined values of Mw were similar (within 12%). For PPDLmolecular weight values reported herein, light scattering Mw andGPC Mw/Mn values were used (see Table 1).

2.5. Fractionation of PPDL sample 5

To separate low molecular weight material in sample 5, corre-sponding to GPC peaks observed at longer retention times, thesample was fractionated as follows. A solution containing 20 mL ofchloroform and 0.2 g of Sample 5 was prepared. The mixture wasmaintained at 20 �C for 24 h with continual mixing by magneticstirring. The ‘insoluble’ PPDL fraction was separated by filtrationand solvent was removed in a vacuum oven at 40 �C for 7 days,giving 0.17 g of ‘insoluble’ solid sample. Solvent removal from the‘soluble’ PPDL fraction was carried out by rotor-evaporation andsubsequent drying in a vacuum oven, where the recovered ‘soluble’PPDL sample was 0.03 g.

2.6. Tensile testing

Dumbbell shaped sample bars with dimensions of 20.0 mm(length)� 4.0 mm (neck width)� 1.5 mm (thickness) wereprepared by press-molding at 130 �C and subsequent quenching atambient temperature. An Instron 5542 tensile testing machine witha 500 N load cell was used for mechanical study (the crossheadspeed was 3 mm/min and the test temperature was 25 �C). TheMerlin software was used to collect and analyze the tensile results(stress was calculated according to the initial cross-section area).The values of tensile strength, Young’s modulus, elongation at yieldand break, stress at yield were obtained by averaging the dataobtained from more than 4 specimens.

2.7. Dynamic mechanical analysis (DMA)

PPDL samples were molded into rectangular bars with dimen-sions of 30 mm (length)� 5 mm (width)� 1.5 mm (thickness).DMA measurements were performed in single-cantilever bendingmode using a dynamic mechanical thermal analyzer (DMTA) (Tritec2000 DMA, Triton Technology Company). Measurements wereperformed from 30 �C to 95 �C at a heating rate of 2 �C/min andfrequency of 1 Hz. Two identical specimens with the same molec-ular weight were evaluated and results reported were mean values.The Triton Technology DMA Software was used to acquire andprocess the data.

J. Cai et al. / Polymer 51 (2010) 1088–1099 1091

2.8. Differential scanning calorimetry

DSC measurements were performed using a differential scan-ning calorimeter (Model 2920, TA Instruments). Temperature cali-bration was carried out using an indium standard. Measurementswere performed under a nitrogen atmosphere at a flow rate of50 mL/min. Typical parameters for experimental measurements areas follows: i) sample cooled to 10 �C, ii) heated to 150 �C at 10 �C/min, iii) held at 150 �C for 3 min and then iv) cooled to 10 �C at 10 �C/min. When temperature reached 10 �C, the sample was heated againto 200 �C at 10 �C/min. The melting temperature, melting enthalpy,crystallization peak temperature, and crystallization enthalpy wereanalyzed by using the TA Universal Analysis software.

2.9. X-ray measurements

Wide-angle X-ray diffraction (WAXD) and small-angle X-rayscattering (SAXS) experiments using the dumbbell shaped bars fortensile testing were carried out at the X27C beam line at theNational Synchrotron Light Source (NSLS), Brookhaven NationalLaboratory (BNL). The wavelength of synchrotron radiation was1.371 Å. A three-pinhole collimation system was used to reducebeam size to 0.6 mm in diameter. Two-dimensional (2D) WAXDand SAXS patterns were collected using a MAR CCD X-ray detector(MAR-USA), which had a resolution of 1024�1024 pixels (pixelsize¼ 158.44 mm). The typical image acquisition time was 30 s foreach data frame. Sample-to-detector distance was 1923.7 mm forSAXS (calibrated by a silver behenate, AgBe, standard) and116.4 mm for WAXD (calibrated by an aluminum oxide, Al2O3,standard). All X-ray images were corrected for background scat-tering, air scattering and beam fluctuations. The obtained 2D SAXSand WAXD patterns were analyzed using the POLAR software toobtain one-dimensional SAXS and WAXD profiles. One-dimen-sional WAXD profiles were then processed by the linear squaremethod to obtain crystallinity.

3. Results

3.1. PPDL synthesis

Previous work by our laboratory demonstrated that polymeriza-tion of PDL for 2 h at 70 �C in dry toluene (PDL to toluene 1:2 wt/vol)with 10 w/w-% monomer-to-catalyst gave PPDL (without fraction-ation) with Mn 7900 g/mol (GPC relative to polystyrene) [17]. Prepa-ration of PPDL with higher molecular weights requires increasingdiffusion constraints that slow chain propagation and decreasingreaction water content to start fewer chains [21,24–26]. To preparea series of PPDL samples (1–3, Table 1) with relatively low molecularweight, the reaction temperature was maintained at 70 �C. Further-more pre-drying of the reaction flask, enzyme and solvent was notcarried out, while mixing was performed by magnetic stirring. After1 h of reaction, the viscosity of the reaction mixture was sufficientlyhigh which stopped magnetic stirring. Hence, the diffusion monomerin the reaction system became slow, which hindered the increase ofmolecular weight [17]. The molecular weight of the product withinthe series of samples 1–3 was only varied by the reaction time. Byincreasing the reaction time from 6 to 13 and 24 h, Mw of PPDL afterprecipitation increased from 2.5 to 4.5 and 8.1�10�4 g/mol (deter-mined by light scattering, see Experimental Section), respectively.

Higher molecular weight PPDL samples 4, 5 and 6 wereprepared by increasing the reaction temperature to 85 �C andusing an overhead stirrer, both with the intent of increasingdiffusivity of reactants. In addition, the reaction flask, syringes forreactant transfer, PDL, toluene and Novozym 435 were all dried

prior to reactions (see Experimental Section). By increasing thereaction time from 8 to 16 and 26, samples 4, 5 and 6 wereprepared having Mw values (�10�4, determined by light scattering)of 18.8, 28.0 and 48.1 g/mol, respectively. For the above, PPDL wassynthesized in quantities up to 40 g and yields from 75 to 80% afterprecipitation.

3.2. Tensile testing

Dumbbell shaped sample bars were prepared (see ExperimentalSection, above) for tensile testing. Sample 1 (Mw 2.5�104) wasfound to be too brittle to be tested. Stress–strain curves for Samples2–6 are illustrated in Fig. 1(a–e). Fig. 1f displays four regionsaccording to slope change in the stress–strain curve. Manson et al.[27] assigned these four regions as: I – linear and nonlinear visco-elasticity, II – neck region/strain softening, III – plastic flow, and IV –strain-hardening. The Young’s modulus was determined as theslope of stress–strain curve at strain below 1% using the linear leastsquare method. The values of strain and stress at yield were the (x,y) coordinates at which the first derivative of the stress–straincurve was zero. Values of elongation and stress at break repre-sented the (x, y) coordinates after which the slope of the stress–strain curve became negative. The averaged results from replicatedexperiments are summarized in Table 2. Inspection of PPDLsamples before stretching and after fracture showed that thevolume of the bar subjected to high strain was found to be abouttwo times that of the original volume (i.e. 50% polymer and 50%void formations). This observation is in agreement with that byWard [28]. Since the stress value in Fig. 1 was estimated by the loaddivided by the initial cross-sectional area, for samples 4, 5, and 6,the true stress values at break were recalculated using the value ofstress at break (in Fig. 1) multiplied by a factor (1þ strain)/2. Forsamples at lower strains (samples 2, 3), the true stress values atbreak were recalculated by the value of stress at break multipliedby a factor (1þ strain), i.e., neglecting the effect of volume increaseat lower strains. The recalculated values of true stress at break arelisted in Table 2. Variations of Young’s modulus, true stress at break,elongation at break, and strain and stress at yield as a function ofPPDL molecular weight are shown in Fig. 2(a–e).

The properties shown in Figs. 1 and 2 and Table 2 reveal unam-biguously that they are dependent on PPDL molecular weight. Fig.1aillustrates that, for sample 2 (Mw¼ 4.5�104), stress first increasedrapidly with strain but then the slope of stress–strain curve began todecrease. When strain reached 4.5� 0.8%, the sample fractured (thefracture was homogeneous as is typical for brittle fracture), while noneck was observed before fracture. As Mw increased to 8.1�104, theshape of the stress–strain curve (Fig.1b) deviated substantially fromthat of sample 2. For sample 3, necking and plastic flow wereobserved during stretching; stress first reached the maximum valueand then decreased, followed by stable propagation before fracture.When the sample bar length reached 237�25% of the initial length,the sample bar fractured. Thus, with an increase in the molecularweight of about 2 times from sample 2 to sample 3, the elongation atbreak increases almost 30 times and the fracture fashion of PPDLchanges from brittle-to-ductile. In addition, a stress-whiteningphenomenon was observed during stretching of sample 3, which isquite different from sample 2. Interestingly, for PPDL ofMw¼ 18.9�104 (sample 4), the elongation at break increased to650� 30% while the strain-hardening phenomenon was alsoobserved (Fig. 1c). With further increases in Mw, samples exhibiteda similar stress–strain curve shape as that of sample 4 and values ofelongation at break remained almost constant (e.g. 700� 70% forMw¼ 28� 104 and 580� 30% for Mw¼ 48.1�104). This suggeststhat for PPDL, it is not necessary to synthesize polymers withmolecular weights above Mw¼ 18.9�104 to attain higher draw

0

5

10

15

20

25

30

Stre

ss

Strain,%

0 50 100 150 2000 1 2 3 4 5 6

0

5

10

15

20

25

Stre

ss

Strain,%

0 100 200 300 400 500 600 700

0

4

8

12

16

Stre

ss

Strain,%

0 200 400 600 800

0

6

12

18

Stre

ss

Strain,%

0 200 400 600

0

6

12

18

Stre

ss

Strain,%

IVIIIIII

Stre

ss

Strain

a b

d

fe

c

Fig. 1. Stress–strain curves for PPDLs with different molecular weights (a,2; b,3; c,4; d,5; e,6). Scheme f shows regions within the stress–strain curve of PPDL samples undergoingstretching. For each sample, repeated stress–strain measurements on fresh PPDL samples are plotted.

J. Cai et al. / Polymer 51 (2010) 1088–10991092

ratios since the elongation at break remains almost constant aboveMw¼ 18.9�104.

The plot of true stress at break (tensile strength) versus Mw

(Fig. 2b) showed a similar trend as that of elongation at break versusMw. The stress at break first increased from 21.3 MPa to 60.8 MPa asMw increased from 4.5�104 to 18.9�104 and then remained almostconstant at high Mw. The curve of Young’s modulus versus Mw

(Fig. 2c) exhibited a minimum at Mw¼ 28.0�104. Young’s modulusfirst decreased from 690 MPa to 290 MPa with increasing Mw, andthen increased to 390 MPa. However, the corresponding curve ofstrain at yield versus Mw first increased with Mw from 12.0 to 20%and then decreased slightly to 17.4%. The trend of stress at yieldversus Mw is almost opposite to that of strain at yield versus Mw,since the stress first decreased from 24.1 to 13.3 MPa, and thenincreased to 18.2 MPa with increasing Mw.

3.3. Thermal analysis

Thermal analysis of several samples with different molecularweight was carried out after processing samples into dumbbellshaped bars prior to tensile testing. Fig. 3displays DSC curves fromfirst heating (Fig. 3a), cooling (Fig. 3b) and second heating (Fig. 3c)scans of PPDL samples (1–6). The trend of peak melting tempera-ture with Mw in heating is similar to that in cooling curves. Valuesof melting enthalpy (ME) and melting temperature (MT) duringfirst and second heating scans, and values of crystallizationenthalpy (CE) and crystallization peak temperature (CPT) duringcooling scans are summarized in Table 3. Since the trends of changein MT and ME during the first heating scan are similar to thoseduring the second heating scan, only values of ME and MT in firstheating and values of CE and CPT in cooling are plotted in Fig. 4a,b.

Table 2Young’s modulus, correlation coefficient in obtaining young’s modulus, elongation to break, stress at break, stress at yielding, and strain at yielding for PPDL samples ofdiffering molecular weight.

Sample No. Mw (�10�4) Young’s Modulus (MPa) Correlation coefficient Elongation at break (%) Stress atbreak (MPa)

Stress atyield (MPa)

Strain atyield (%)

2 4.5 690� 40 0.9995 4.5� 0.8 20.4� 3.2(21.3� 3.3)a

– –

3 8.1 620� 50 0.99934 137� 25 17.7� 1.3(42.0� 3.0)

24.1� 2.0 12.0� 1.0

4 18.9 450� 20 0.99872 650� 28 16.2� 0.5(60.8� 2.0)

17.2� 0.3 13.0� 1.0

5 28.0 290� 30 0.99865 703� 72 13.7� 1.0(55.0� 4.0)

13.3� 1.3 20.6� 1.3

6 48.1 390� 10 0.9989 580� 30 17.3� 0.9(58.8� 3.1)

18.2� 0.3 17.4� 0.4

a The true stress at break calculated after cross-section area correction.

J. Cai et al. / Polymer 51 (2010) 1088–1099 1093

Fig. 4a indicates that ME first increased from 134.0 J/g to 164.4 J/gand then decreased afterward with increasing Mw. When Mw was28.0�104, ME reached a minimum value (101.6 J/g); when Mw was48.1�104, ME climbed back up to 115.3 J/g. Similarly, MT firstincreased from 97 �C to 103.2 �C, and then decreased to 91.7 �C, andagain increased to 99.4 �C with increase of Mw. Upon crystallizationof PPDL during cooling, the changes of CE and CPT with Mw weresimilar to those of ME and MT (Fig. 4b). Generally, the value of CEwas smaller than that of ME in the same sample, possibly due torecrystallization upon heating [29].

3.4. X-ray measurements

To compare DSC results, wide-angle X-ray diffraction (WAXD)experiments were performed on the same dumbbell shapedsample bars prior to tensile testing. Fractions of crystalline phasesplotted in Fig. 5 were obtained by peak deconvolution of integratedWAXD profiles. The trend of crystallinity change with Mw wassimilar to the variation of ME and CE with Mw. To compare the DSCand X-ray results, the values of crystallization enthalpy from cool-ing (instead of the melting enthalpy) were used in order to elimi-nate the effect of recrystallization during DSC scanning. Theequilibrium melting enthalpy of PPDL had been estimated to be264 J/g and 233 J/g, respectively, by Wunderlich [29] and Lebedvet al. [30]. If 233 J/g was used to calculate crystallinity values fromDSC results, calculated crystallinity values would be higher thanthose from WAXD for samples with Mw less than 44.0�104 whilethe results were reverse when Mw was above 44.0�104. If 264 J/gwas used, the calculated crystallinity from DSC was slightly higherthan that from WAXD for samples with Mw less than 18.9�104;however, the calculated crystallinity from DSC was substantiallylower than that from WAXD when Mw was above 18.9�104.Considering that the crystallinity calculation from WAXD includescontributions of both crystalline and interphase regions, whilecrystallization enthalpy from DSC is only from the crystal forma-tion, the crystallinity value estimated from WAXD should be higherthan those from DSC if the latter does not involve recrystallization.Thus, the value of 264 J/g was used in this study for calculations ofPPDL crystallinity from DSC.

3.5. Dynamic mechanical analysis (DMA)

DMA was used in previous work [18] to measure the glasstransition (�27 �C) and the effect of water content on b and grelaxations for PPDL with Mn¼ 6.5�104. In this study, DMA wasused to compare the storage modulus of PPDL with different Mw

values in the temperature range from 30 �C to 95 �C. Since sample 1was brittle, only samples 2–6 were subjected to DMA tests. Fig. 6aillustrates curves of storage modulus vs. temperature for PPDLsamples with different molecular weights. For each sample, thestorage modulus decreased with increasing temperature and nodistinct transitions were observed. The results in samples 3 and 4exhibited an intersection at 85 �C, while samples 5 and 6 exhibitedan intersection at 56 �C. Storage moduli at 40–90 �C for fivesamples are listed in Table 4 and displayed in Fig. 6b. This figureindicates that the trend of storage modulus changes with Mw at40 �C was similar to that of the Young’s modulus with Mw (Fig. 2c).Interestingly, as the temperature increased from 40 to 60 �C, thediscrepancy between samples 5 and 6 decreased. For temperaturesabove 60 �C, the storage moduli of samples 5 and 6 are the same.This implies that above 60 �C, the molecular networks formed forsamples 5 and 6 have the same elastic rigidity.

4. Discussion

This section considers the effects of Mw on PPDL mechanical,thermal and crystallization properties and compares these to thoseof linear PE.

4.1. Effects of molecular weight on PPDL properties

The effect of molecular weight of linear HDPE on its tensileproperties was studied by Ward et al. [22]. They reported thatYoung’s modulus decreases with increasing Mw. Since PPDL hasa similar chemical structure as linear PE, its Mw dependence ofYoung’s modulus should show a similar trend. However, The resultsherein appear to be different from those reported by Ward et al.[22] For example, in Fig. 2c, Young’s modulus first decreases toa minimum (at sample 5, Mw 28.0�10�4) and then increases.Without sample 5, behavior of PPDL would be similar to linear PE.Perhaps this can be rationalized as follows. Several other studieshave indicated that the degree of crystallinity is the primary factoraffecting the Young’s modulus of semi-crystalline polymers[31–34]. It was known that low Mw fractions can act as diluents andretard the crystallization of higher Mw fractions [29]. Furthermore,low Mw fractions usually contribute less to the Young’s modulusdue to the lower possibility to form entanglements or tie moleculesto transmit the stress [34]. In sample 5, the GPC spectrum clearlyexhibits several peaks corresponding to low Mw fractions (Fig. 7).These low Mw fractions would be expected to decrease crystallinityand lower the Young’s modulus. To verify this hypothesis, thefollowing experiment was performed. Sample 5 was first immersed

0 1x105 2x105 3x105 4x105 5x105

0

200

400

600

800a b

dc

e

Elo

ng

atio

n at

Break,%

0 1x105 2x105 3x105 4x105 5x1050

20

40

60

Stress at B

reak (M

Pa)

Mw

Mw

Mw

1x105 2x105 3x105 4x105 5x10510

15

20

25

Stra

in a

t Y

ield

(M

Pa)

1x105 2x105 3x105 4x105 5x105

12

16

20

24

28

Stress at Y

ield

(M

Pa)

Mw

0 1x105 2x105 3x105 4x105 5x105200

300

400

500

600

700

800

su

lu

do

M

s'g

nu

oY

)apM(

Mw

Fig. 2. Variations of as a function of PPDL molecular weight for tensile tests at 25 �C of: (a) elongation at break, (b) true stress at break, (c) young’s modulus, (d) strain at yielding and(e) stress at yielding. The solid and dotted lines in c indicate how trends in Young’s modulus as a function of molecular weight changes with and without sample 5, respectively.

J. Cai et al. / Polymer 51 (2010) 1088–10991094

in chloroform for 24 h at 20 �C to solubilize a large part of low Mw

fractions. The remaining insoluble material after thorough removalof solvent was subjected to GPC and DSC analyses. The GPC chro-matogram displayed in Fig. 7 indicates that a substantial part of lowmolecular weight fractions were removed by solubilization inchloroform. DSC results indicate that, after solvent extraction, themelting peak of sample 5 shifted to a higher temperature duringboth first and second heating scans (Fig. 8). The melting tempera-ture, melting enthalpy for the first and second heating, the coolingcrystallization peak, and cooling crystallization enthalpy are listedin Table 3 for comparison with data for non-fractionated sample 5.Indeed, the value of melting enthalpy increases after extraction andfalls between those of samples 4 and 6. Thus, it appears that thecrystallinity and Young’s modulus of PPDL have a similar molecularweight dependence as linear PE based on studies by Ward et al.[28], i.e., both decrease with increasing Mw.

4.2. Young’s modulus as a function of crystallinity

Since crystallinity plays a major role to influence Young’smodulus, their relationship was investigated. The crystallinity wascalculated by dividing the cooling crystallization enthalpy with theequilibrium enthalpy value of 264 J/g [29]. The relationshipsbetween crystallinity and Young’s modulus are shown in Fig. 9a.The data set can be represented by a straight line without theextrapolation to the origin. This behavior was also observed byMandelkern et al. [35], who measured the crystallinity of frac-tionated linear PE by density measurement. Flory and Yoon [35]pointed out that the initial chain topology in the molten polymercan be conserved during crystallization. In other words, the topo-logical features such as entanglements, knots, loops and relatedstructures can be regulated in the inter-lamellar amorphous region.Without question, these types of structures will affect the elastic

6

5

4

3

2

1

Hea

tflow

End

ow D

own

6

5

4

3

2

1

Hea

tflow

End

ow D

own

60 80 100 120 140

40 50 60 70 80 90 100 110 120

60 80 100 120 140

6

5

4

3

2

a

b

c

1H

eatfl

ow E

ndo

Dow

n

Temperature

Temperature

Temperature

Fig. 3. DSC thermograms recorded for PPDL samples 1– 6 during: (a) first heating,(b) cooling and (c) second heating.

Table 3DSC results of melting point (Tm), crystallization peak temperature (Tcp), meltingenthalpy (DHm), and cooling crystallization enthalpy (DHc) for PPDL of differing Mw

values.

Mw� 10�4 First heating Second heating Cooling

DHm, J/g Tm, �C DHm, J/g Tm, �C DHc, J/g TcP, �C

2.5 134.0 97.0 131.4 94.7 134.1 76.14.5 164.4 100.2 151.0 95.0 141.1 77.78.1 147.3 103.2 143.1 96.0 135.6 77.618.9 120.7 97.2 128.4 94.6 118.0 75.428.0 101.6

(118.6)a91.7(103.2)

99.3(115.8)

91.4(98.3)

92.2(103.8)

68.5(70.5)

48.1 115.3 99.4 110.1 97.1 97.7 71.2

a Samples extracted with chloroform for 24 h at 20 �C.

0 1x105 2x105 3x105 4x105 5x105

90

100

110

120

130

140

150

TcpH,

J/g

Mw

60

65

70

75

0 1x105 2x105 3x105 4x105 5x105

100

110

120

130

140

150

160

170a

b

H, J

/g

Mw

90

100

Tm

Fig. 4. Plots of melting enthalpy, melting point (Tm), crystallization enthalpy andcrystallization peak temperature (Tcp) as a function of PPDL molecular weight recordedby DSC during: (a) first heating and (b) cooling.

J. Cai et al. / Polymer 51 (2010) 1088–1099 1095

properties. In addition, the interphase region that connects theordered crystalline region and the isotropic conformational disor-dered amorphous region can also contribute to elastic modulus[34]. A schematic diagram indicating different chain topologies inthe lamellar structure of PPDLs with different Mw is shown inFig. 10, which will be used to explain the different observed prop-erties in the samples.

4.3. Yield stress as functions of crystallinity and crystallite thickness

The plot of yield stress versus crystallinity is shown in Fig. 9b,where the behavior can be rationalized by the occurrence of one orboth of the following phenomena [36–38]. One possibility isa partial melting-recrystallization process is induced by deforma-tion [36]. An alternative explanation is based on the screwdislocation theory [37,38]. The mechanism of deformation-induced

0 1x105 2x105 3x105 4x105 5x10540

45

50

55

Wc,%

Mw

Fig. 5. Crystallinity of PPDL, determined by wide-angle X-ray diffraction (WAXD), asa function of PPDL molecular weight.

0 1x105 2x105 3x105 4x105 5x1050.0

2.0x108

4.0x108

6.0x108

8.0x108

Stor

age

Mod

ulus

, Pa

Mw

40°C0°C

0°C0°C0°C0°C

5 6 7 8 9

40 50 60 70 80 900.0

3.0x108

6.0x108

6

5

4

3

2a

b

Stor

age

Mod

ulus

Temperature

Fig. 6. (a) Plots of storage modulus as a function of temperature for PPDL withdifferent molecular weight (2, Mw¼ 45,000; 3, Mw¼ 81,000; 4, Mw¼ 189,000; 5,Mw¼ 280,000; 6, Mw¼ 481,000); (b) dependence of storage modulus on PPDLmolecular weight at selected temperatures.

Table 4Storage modulus of PPDL samples of differing Mw at varying temperatures.

Mw� 10�4 Modulus, Pa

40 �C 50 �C 60 �C 70 �C 80 �C 90 �C

4.5 8.0� 108 6.8� 108 5.5� 108 4.4� 108 3.4� 108 2.6� 108

8.1 6.6� 108 5.4� 108 4.3� 108 3.4� 108 2.6� 108 1.8� 108

18.9 5.6� 108 4.8� 108 3.9� 108 3.2� 108 2.5� 108 1.9� 108

28.0 4.0� 108 3.4� 108 2.7� 108 2.1� 108 1.6� 108 1.1� 108

48.1 4.5� 108 3.5� 108 2.7� 108 2.1� 108 1.6� 108 1.2� 108

J. Cai et al. / Polymer 51 (2010) 1088–10991096

melting-recrystallization is based on previous experimental resultsby observation of transmission electron microscopy [39] and small-angle neutron scattering [40] results with linear PE. The screwdislocation theory argues that the shear yield stress (i.e., experi-mental yield stress/2) is related to the crystallite thickness by thefollowing expression:

sy ¼k

4p

"exp

2pDGc

kb2Lcþ 1

!#�1

(1)

where sy is the shear yield stress, Lc is the crystal lamellar thickness,k is a function of the crystal shear modulus, b is the Burgers vectorhaving the same value as the PPDL unit cell c-axis, DGc is the criticalactivation energy for dislocation growth, with values in the rangebetween 40 and 80 kT. To correlate the relationship between yieldstress and crystallite thickness, SAXS experiments were carried outwhere the results (the Lorentz-corrected SAXS profiles) are shownin Fig. 11. These profiles exhibit two peaks at very low q values. Toobtain the long period L and crystallite thickness Lc, the one-dimensional correlation function K(r) is calculated by the followingequation [41]:

KðrÞ ¼ 12p2

Zp

0

Iq2 cos ðqrÞdq (2)

where I(q) is the scattering intensity and r is the distance for whichthe electron density correlation is measured. The long periods forsamples 3, 4, 5 and 6 were 17.5 nm, 18.9 nm, 19.3 nm and 18.1 nm,respectively. Assuming the crystal thickness should be thicker thanthe amorphous regions in the two-phase model, the estimated

4 6 8 10

3 2 1

After extraction

Original

Co

ncen

tratio

n sig

nal(M

V)

Elution time(minutes)

Fig. 7. Comparison of GPC curves for PPDL sample 5 before and after solventextraction.

40 80 120

After extraction

Original

Hea

tflow

End

ow D

own

Temperature

40 80 120

After extraction

Original

a

b

Hea

tflow

End

ow D

own

Temperature

Fig. 8. Effect of sample fractionation melting of PPDL sample 5 determined by DSCthermograms during: (a) first heating and (b) second heating scans.

0.0 0.2 0.4 0.6 0.8 1.00

200

400

600

800a

b

Yo

un

g's

M

od

ulu

s

Xc

Xc

0.0 0.2 0.4 0.6 0.8 1.00

5

10

15

20

25

30

Stress at Y

ield

Fig. 9. Plots of Young’s modulus (a) and stress at yield (b) as a function of PPDLcrystallinity.

amorphous

crystalline

interfacial

interfacial

crystalline

tight-fold

tie-chain

entanglement

i

a

c

i

c

a b

Fig. 10. Possible mechanism responsible for the brittle-to-ductile transition in: (a)lower molecular weight fraction and (b) higher molecular weight fraction.

J. Cai et al. / Polymer 51 (2010) 1088–1099 1097

crystalline thickness values are 10.7 nm, 12.2 nm, 13.4 nm, 11.1 nm,respectively. The close values of the crystalline thickness lead us toconclude that the screw dislocation theory is not appropriate toexplain yield stress data.

4.4. Strain and true stress at break

Several studies [22,35,42–45] have been reported dealing withthe molecular weight effect on draw ratio for PE. The results indi-cate that the draw ratio decreases (e.g. 18–3) with increasingmolecular weight of linear PE, whereas the true stress at breakvalue exhibits a maximum at Mw¼ 30.0�104 [33]. Results for PPDLsamples are different from that obtained with PE. For PPDL, whenMw increases to 18.9�104, the draw ratio reaches an asymptoticvalue of about 7.5 (Fig. 2a). Interestingly, the corresponding truestress at break also reaches an asymptotic value of about 60.0 MPaat Mw¼ 18.9�104 (Fig. 2b). One possible explanation for thediscrepancy is as follows. Since the PPDL chain consists of C–C andC(]O)–O (or–O–C–) bonds along the backbone, where the C–Cbond has larger bond energy than C(]O)–O (or –O–C–) bonds.Hence, the stress distribution should be heterogeneous along thechain during stretching. In high Mw samples (>18.9�104),

0.1 0.2

5#

6#

4#

3#

q(nm-1

)

q2I(q

) (a

.u

.)

Fig. 11. Lorentz-corrected SAXS profiles of PPDL samples 3– 6 before stretching.

J. Cai et al. / Polymer 51 (2010) 1088–10991098

increased Mw would lead to enhanced strength of the entangle-ment network in the molten melt as well as in the inter-lamellaramorphous region when solidified. However, as the C–O bond isweaker than the C–C bond, the former probably breaks at a lowerstress under deformation, thus limiting the ultimate strength ofPPDL. This hypothesis is consistent with the observation of constantstress as well as the constant elongation at break value aboveMw¼ 18.9�104 under deformation. This may also explain why PEhas a much higher draw ratio (18 or higher) than PPDL of the sameor higher Mw, as the PE chain has only carbon–carbon bonds [32]. Ingeneral, variations of elongation and true stress at break (tensilestrength) with Mw observed in this study are consistent with thosereported in a tensile study of cellulose acetate by Sookne et al. [45]and the theoretical predication of Flory [46].

4.5. Brittle-to-ductile transition

The brittle-to-ductile transition in semi-crystalline polymers canbe affected by many factors such as deformation rate and crystal-linity [28,34]. Usually, increases in deformation rate and crystal-linity will cause the brittle-to-ductile transition to shift towardshigher Mw. However, molecular weight is still the most importantfactor that affects the material if the brittle-to-ductile transition isobserved. Mandelkern et al. [35] found that for Mw� 4.0�104, PEmaterials are brittle over the accessible range of crystallinities.When Mw is 7.0�104, the brittle-to-ductile transition can beobserved. Furthermore, when the crystallinity is higher, the tran-sition becomes sharper. In higher Mw PE (about 31.6�104) thebrittle-to-ductile transition was not observed by increasing PEcrystallinity. For PPDL, the Mw at which the brittle-to-ductile tran-sition occurs (Fig. 2) is in agreement with the Mw range observed forPE. In addition, quenched PPDL sample 3 exhibited ductile behavior,while the annealed sample with higher crystallinity revealed brittlebehavior. This finding agrees with Mandelkern’s experiments onlinear fractioned PE with Mw¼ 7.0�104 [35].

Finally, we caution that when assessing the mechanical prop-erties of PPDL samples of different molecular weights, in addition tocrystallinity, crystallite thickness, inter-lamellar thickness anddensity of physical entanglements in the amorphous region, thefraction of the interfacial region and supermolecular morphologyshould also play an important role [34,47,48], which will beconsidered in our future studies.

5. Conclusions

Synthesis of large-scale PPDL samples was performed by lipasecatalysis. Variation of PPDL Mw from 2.5 to 48.1�104 was achievedby manipulating reaction variables including reaction watercontent, method of mixing and reaction time. Tensile testing, DSC,X-ray, DMA and GPC were used to investigate the effect of molecularweight on mechanical, thermal and crystalline material properties.Cold-drawing tensile tests at room temperature revealed a brittle-to-ductile transition for PPDL samples with Mw values between4.5�104 and 8.1�104. For PPDL with Mw¼ 8.1�104, the entan-glement network strength in non-crystalline regions is notsufficiently high to transmit stress during stretching, whereby inter-fibrillar slippage dominates until fracture. As PPDL Mw is increasedfrom 18.9�104 and above, the entanglement network strength isgreatly enhanced and strain-hardening takes place at high strainsprior to failure. Furthermore, the elongation at break and tensilestrength (i.e., true stress at break) reach asymptotic values of 650%and 60.8 MPa, respectively. The trends for changes in Young’smodulus, melting enthalpy (from DSC) and crystallinity (fromWAXD) as a function of molecular weight are all similar. Abnor-mality of one sample (5) was explained by its contamination withhigh levels of a lower molecular weight fraction. Storage modulus(from DMA) revealed similar molecular weight dependence trendsat temperatures below 60 �C. However, above 60 �C, storage moduliof higher molecular weight samples becomes indistinguishable,indicating that the crystalline network structures of these samplesat higher temperatures are also similar. Overall comparisonsbetween PPDL and linear high density polyethylene (HDPE)mechanical properties showed similar trends in Young’s moduluswith molecular weight, but differing trends with respect to elon-gation at break and true stress at break as a function of molecularweight. The latter differences are explained by the presence of a lowbut persistent density of C–O bonds in ester links of PPDL that arenot present in PE. We therefore conclude that PPDL-like polyestershave excellent potential to function in similar ways to PE. This bodeswell to the potential further development of similar or relatedpolymers for commercial use. Indeed, our laboratory has developedbiocatalytic methods using an engineered Candida tropicalis strainto convert fatty acids, such as tetradecanoic acid, in volumetricyields of up to 160 g/L to their corresponding u-hydroxyfatty acid(e.g. u-hydroxytetradecanoic acid) (manuscript in review). Subse-quent conversion of u-hydroxyfatty acids to polyesters by conden-sation polymerization will provide a low-cost scalable route tobiobased PPDL-like materials that can function in similar ways to PE.

Acknowledgements

The authors thank the National Science Foundation (NSF) andindustrial members of NSF-I/UCRC for Biocatalysis andBioprocessing of Macromolecules at NYU/Polytechnic Universityfor financial support, intellectual input, and encouragement duringthe course of this research. BH also thanks support from NSF (DMR-0906512). We are also grateful to Mr. Seong Chan Park andDr. Miriam Rafailovich at SUNY at Stony Brook for providing accessto their Instron Tensile testing and DMA equipment.

References

[1] Mazzocchetti L, Scandola M, Jiang ZZ. Macromolecules 2009;42:7811.[2] Ceccorulli G, Scandola M, Kumar A, Kalra B, Gross RA. Biomacromolecules

2005;6(2):902.[3] Jiang ZZ, Azim HS, Gross RA, Focarete ML, Scandola M. Biomacromolecules

2007;8:2262.[4] Keller A, Martuscelli E, Priest DJ, Udagawa Y. J Polym Sci Part A-2 1971;9:1807.

J. Cai et al. / Polymer 51 (2010) 1088–1099 1099

[5] Patel GN, Keller A. J Polym Sci Polym Phys Ed 1975;13(2):303.[6] Haridoss S, Perlman MM. J Appl Phys 1984;55:1332.[7] Sezgin A, Cha WS, Smith JM, Mccoyb J. Ind Eng Chem Res 1998;37(7):2582.[8] Zhang JG, Jiang DD, Wilki CA. Polym Degrad Stab 2006;91(4):641.[9] Fang CQ, Li TH, Zhang ZP, Jing DQ. Polym Comp 2008;29(5):500.

[10] <http://www.rrcap.unep.org/uneptg05/outline/env/white.pdf>.[11] Meulen I, Geus M, Antheunis H, Deumens R, Joosten EAJ, Koning CE, et al.

Biomacromolecules 2008;9:3404.[12] Jedlinski Z, Juzwa M, Adamus G, Kowaczuk M. Macromol Chem Phys

1996;197:2923.[13] Makovetsky KL. Vysokomol Soedin Ser B 1999;41(9):1525.[14] Yevstropov A, Lebedev BV, Kiparisova Y. Vysokomol Soedin Ser A 1983;25:

1679.[15] Zhong Z, Dijkstra PJ, Feijen J. Macromol Chem Phys 2000;201:1329.[16] Uyama H, Takeya K, Kobayashi S. Bull Chem Soc Jpn 1995;68:56.[17] Kumar A, Kalra B, Dekhterman A, Gross RA. Macromolecules 2000;33:6303.[18] Focarete ML, Scandola M, Kumar A, Gross RA. J Polym Sci Part B Polym Phys

2001;39:1721.[19] Simpson N, Takwa M, Hult K, Johansson M, Martinelle M, Malmstrom E.

Macromolecules 2008;41:3613.[20] Gazzano M, Malta VA, Focarete ML, Scandola M, Gross RA. J Polym Sci Part B

Polym Phys 2003;vol. 41:1009.[21] Focarete ML, Gazzano M, Scandola M, Kumar A, Gross RA. Macromolecules

2002;35:8066.[22] Capaccio G, Ward IM. Polymer 1975;16:239.[23] Matthijs de Geus. Enzymatic catalysis in the synthesis of new polymer

architectures and materials. Ph.D. thesis. Eindhoven University of Technology,Netherlands; Feb. 2007 [chapter 5].

[24] Kumar A, Gross RA. Biomacromolecules 2000;1:133.

[25] Bisht KS, Henderson LA, Gross RA, Dl Kaplan, Swift G. Macromolecules1997;30(9):2705.

[26] Mei Y, Kumar A, Gross RA. Macromolecules 2002;35:5444.[27] Manson JA, Hertzberg RW. CRC Crit Rev Macromol Sci 1973;1:433.[28] Ward IM. Mechanical properties of solid polymers, vol. 399. John Wiley & Sons

Pub; 1983 [chapter 12].[29] Wunderlich B. Advanced thermal analysis. Databank of thermodynamic

properties of linear macromolecules and small molecules. Knoxville, TN, USA:Department of Chemistry, University of Tennessee; 1995.

[30] Lebedev B, Yevstropov A. Makromol Chem 1984;185:1235.[31] Keith HD, Paden FJ, Vadimsky RG. J Appl Phys 1971;42:4585.[32] Capaccio G, Crompton TA, Ward IM. J Polym Sci Part B Polym Phys

1976;14:1641.[33] Mandelkern L. Polym J 1985;17:337.[34] Mandelkern L. Acc Chem Res 1990;23:380.[35] Kennedy MA, Peacock AJ, Mandelkern L. Macromolecules 1994;27:5297.[36] Flory PJ, Yoon DY. Nature 1978;272:226.[37] Young RJ. Philos Mag 1974;30:85.[38] Crist B, Fischer CJ, Howard PR. Macromolecules 1989;22:1709.[39] Philips PJ, Philipot RJ. Polym Commun 1986;27:307.[40] Wu W, Wignall GD, Mandelkern L. Polymer 1992;33:4137.[41] Strobl GR, Schneider M. J Polym Sci Part B Polym Phys 1980;18:1343.[42] Andrews JM, Ward I. J Mater Sci 1970;5:411.[43] Williamson GR, Wright B, Haward RW. J Appl Chem 1964;14:131.[44] Popli R, Mandlkern L. J Polym Sci Polym Phys Ed 1987;25:441.[45] Sookne AM, Harris M. Ind Eng Chem 1945;37(5):478.[46] Flory PJ. J Am Chem Soc 1945;67:2048.[47] Peterlin A. J Mater Sci 1971;6:490.[48] Vincent PI. Polymer 1960;1:425.